Excavator Camera Control Using Operator Feedback to Suppress False Detections
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing techniques for detecting objects around excavators using stereo cameras can result in inaccurate detection, leading to inconveniences and decreased work efficiency due to false alarms and unnecessary safety controls.
Innovation Solution
A system that includes a controller and display device capable of correcting false object detections by allowing operators to designate false detections for suppression, using a trained model to identify and store these objects in a database, thereby preventing further false alarms and improving detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If object detection is performed using a stereo camera, then surrounding objects can be monitored, but false detection occurs leading to inaccurate detection results
Solution Approach 1:
The system incorporates a feedback mechanism where the operator can correct false detection results by designating objects as false alarms. The control device stores this correction information and uses it to suppress detection of similar objects in the future, thereby improving detection accuracy through operator feedback.
Solution Approach 2:
The system enables self-correction by allowing the operator to manually designate false detections. The control device automatically learns from these designations and adjusts its detection algorithm to suppress similar false alarms, making the system progressively more accurate without requiring complete redesign.
2Productivity
If false object detections are not suppressed, then detection coverage is maintained, but unnecessary safety controls are triggered decreasing work efficiency
Solution Approach 1:
The system performs preliminary action by storing false detection information and using it to pre-suppress detection of similar objects before they can trigger unnecessary safety controls. This proactive approach prevents inefficiency from occurring in the first place.
Solution Approach 2:
Operator feedback about false detections is stored and used to modify future detection behavior. This feedback loop ensures that unnecessary safety controls are not triggered by similar false alarms, thereby maintaining work efficiency while improving reliability.
3Quantity of substance
If detection sensitivity is increased to catch all objects, then detection coverage improves, but false alarms increase
Solution Approach 1:
The system dynamically adjusts detection sensitivity based on stored false alarm information. By learning from operator corrections, the detection algorithm adapts its sensitivity thresholds to distinguish between real objects and false alarms, maintaining comprehensive detection while reducing false positives.
Solution Approach 2:
The feedback mechanism allows the system to learn from false alarms and adjust its detection parameters accordingly. This enables the system to maintain high detection coverage for real objects while suppressing detection of objects that have been identified as false alarms through operator feedback.
Data Source
AI summary
An excavator includes a lower traveling body; an upper slewing body that is slewably mounted on the lower traveling body; a photographing device that is attached to the upper slewing body; a display device configured to display photographed image data that is obtained by the photographing device; and a control device configured to, in response to receiving an operation to designate an object included in the photographed image data, perform, on information indicating the object, a setting to suppress detection of the object.